Genetic factors play an important role in the etiology of breast cancer, a complex, multifactorial disease. To date, genome-wide association studies (GWAS) have discovered approximately 67 common genetic susceptibility loci for breast cancer risk. However, with the exception of a few loci, all others were identified initially in studies conducted among women of European ancestry. Among the 67 index SNPs reported to date, only about a half of them could be directly replicated in Asians. Given differences in genetic architecture across different ethnic populations, we hypothesize that different risk variants may exist in Asian-ancestry populations in some of the loci in which the index SNPs were not replicated in Asians. Multiple studies have showed that imputation based on the 1000 Genomes Project data provides better chance to identify novel risk variants than that based on the HapMap data since data in the 1000 Genomes Project have much denser SNPs especially low allele frequency SNPs and a larger sample size. Over the past few years, we have genotyped ~9,400 breast cancer cases and controls of Asian ancestry using Affymetrix 6.0 SNP arrays. We propose to impute data for these samples using the most recent 1,000 Genomes Project data as reference to evaluate 10 breast cancer loci in which the index SNPs were not replicated in Asians. Promising SNPs will be further investigated in an independent set of 8,400 cases and controls of Asian ancestry. With strong methodology and very cost- efficient study design, we anticipate that novel genetic variants will be identified in these loci in Asian ancestry populations. These newly-identified variants could significantly improve our understanding of breast cancer genetics and biology and could be used for cancer screening and risk assessment aimed at identifying high- risk women for targeted breast cancer prevention.